Investigate possible sedentism changes through the examination of pithouse measurements.
Pithouse count and archaeological site count data seems to show an increase in population leading into the Middle Jomon period, and dropping towards the end of Middle Jomon heading into late Jomon. The problem with using archaeological sites or pithouse counts as proxies for population is that to do so you need to assume stable mobility patterns. An increase in buildings or sites could be interpreted as an increase in population, but it could also be interpreted as an increase in mobility, as more moves in a year could result in more dwellings. Due to poor preservation of organic remains, faunal or botanical analyses, which are often used to determine seasonality or sedentsim, are difficult to do in Japan. Therefore an alternative method to estimate sedentism could help to elucidate whether increases in pithouses are true increases in population, or if they are instead indicative in an increase in mobility.
Architectural energetics indicate that it is more efficient to build dwelings intended to be used for a longer period of time in a robust manner. This involves a more selective approach to building materials, choosing stronger, more resistant materials and ensuring that the sizes of materials are more standardized. This requires more energy for the initial construction, but requiring less mainenance and energy input long-term. On the other hand, if a dwelling is only intended to be occupied for a relatively short period of time, it is more efficient to use expedient materials that are easier to gather and process. The dwelling won’t resist the elements as well as a sturdier structure, but as the intended use period is short, this is less of a factor.
By measuring the size of pithouses and related postholes from the Tama New Town area, I will investigate whether there were any changes in pithouse size or in posthole size and variation that might indicate a change in residential mobility.
In the Tama New Town area there was an increase in pithouses from the start of the Middle Jomon period apx 5400 calBP to apx 4800 calBP. This was followed by a decrease in pithouse numbers throughout the rest of the Middle Jomon period, ending at 4400 calBP. There appears to be a slight plateau from 5200 calBP to 5000 calBP followed by a steeper increase from 5000 to 4800 calBP.
While investigating Tama New Town, I broke the types of pithosues into 5 main groups: Pit, Wall, Main, MainWall, and Mirror. The number displayed is the total of all 1000 MCMC runs, not actual total numbers for the site.
Pit types showed no indications of inner main posts or surrounding wall posts.
These were relatively rare
Wall types lacked inner main posts, but did have outer wall posts. These make up a small percentage of the pithouses, but the number appears to have increased from 5100 calBP towards the end of the period where the number decreases slightly.
Main types were one of the major pithouse types. The are identified by the presence in inner main posts, but lack the exterior wall posts. The number of Main types increased from the start of the Middle Jomon period until apx 4800 calBP where they experienced a two phase drop heading towards the end of the period.
MainWall types posessed both inner main posts as well as a ring of outer wall posts. MainWall types increased from apx 5400 to 5100 calBP where they show a sharper increase heading until apx 4800 calBP. From that point they show a similar drop as the Main types heading towards the end of the period.
Mirror types were named due to their handmirror type floorplan, where a long entryway was attached to the main dwelling portion. These types also consisted of stone paved floors, something that the other types do not exhibit. The first Mirror types started apx 5000 calBP in Tama New Town. The number of Mirror types increased in the second half of the Middle Jomon period and experienced a slight decrease at the end of the period. At the end of the Middle Jomon there were more Mirror types than Main or Mainwall types in Tama New Town.
From the beginning of the Middle Jomon period to apx 4800 calBP, the average size of pithouses increased from apx 15m2 to 20m2, followed by a decrease in size. During that same time period, pithouse length increased from apx 4.25m to 5m, followed by a decrease back to the 4.25m size that started the period. Edge effects obscure the beginning and end of the period, but the trend of pithouse size increase and decrease is clear, and it corresponds with the number of pithouses.
While trends in pithouse size were generally shared between the different pithouse types, there were some differences.
Pit types displayed a large degree of uncertainty, but there appears to be an increasing trend in pithouse size from apx 3.5m starting around 5300 calBP curving upwards to apx 4.5m at 4600 calBP. Floor space increased from apx 10m2 at the start of the period to apx 15m2 at the end of the period.
#### Wall Types
Wall types showed the greatest degree of uncertainty showing significant fluctuations. Temporal uncertainty is clear from length regressions, showing different start dates depending on the MCMC run. No clear increasing or decreasing trend is present, but no real conclusion can be drawn due to the clear degree of uncertainty.
Main pithouse type trends looked similar to pit style trends. From 5300 to 4800 cal BP average pithouse size increased from approximately 4.5m to 5m, or 15m2 to apx 20m2, and then decreased back to apx 4.5m or 15m2 or slightly smaller heading towards the end of the period.
Somewhat surprisingly, MainWall types displayed a somewhat different trend. Mainwall types show a gradual increase in size throughout the Middle Jomon period, starting at apx 4.5m and ending at apx 5.5m long, and a floor space measurement of slightly over 15m2 increasing to over 20m2. Like all the other plots here, the start and end of the period shows some uncertainty due to edge effects.
Due to the small sample size and relatively short temporal range, Mirror type length and floor space regressions displayed high degrees of uncertainty and are not of much diagnostic value.
I expected posthole depth to track with pithouse size, but that doesn’t appear to be the case. Mean posthole depth is fairly flat throughout the Middle Jomon period. Posthole depth only showed an R of 0.32 compared against floor space, and a 0.31 correlation with pithouse length.
Posthole diameter did show an increase in size heading into the middle of the period and decreased from apx 4900 calBP to the end of the period. Both the beginning and the end of the periods show a high degree of uncertainty, likely due to edge effects. Posthole diamter was more closely correlated with pithouse size, with an R of 0.63 compared to floor space and 0.64 compared to pithouse length.
As posthole depth remained largely unchanged throughout the Middle Jomon period, posthole volume trends tracked closely with posthole diameter trends. Posthole volume and posthole diamter had a correlation coefficient of 0.953. Volume was the posthole measurement most closely correlated with pithouse floor space or length, but was still only moderately correlated with an R of 0.64 and 0.65 respectively.
These correlation matrices display all three posthole measurement correlations at once. In the left figure pithouse floor space measurements are compared to mean posthole measures. In the right figure pithouse length measurements are compared to mean posthole measures.
These findings show that while posthole diameter and volume appear to show strong correlations with pithouse length and floorspace overall, the measures of posthole depth seems fairly independent of pithouse size in general. Outside of the degree of posthole volume variation, which shows a modest correlation with pithouse size, measures of posthole depth and diameter variation show little correlation with pithouse size, meaning the selection and standardization of posts sizes are independent of pithouse size.
Posthole depth variation appears to have decreased from apx 5300 calBP to 4900 calBP. From 4900 calBP to apx 4600 calBP posthole depth variation appears to have increased. The beginning and end of the period have much higher degrees of uncertainty, most likely the result of edge effects. The degree of posthole size variation measured in the standard deviation of posthole measures showed even less correlation than the posthole mean measures. Posthole depth SD had a correlation of 0.108 to floor space and 0.0825 correlation to pithouse length.
Posthole diameter variation appears to have increased from 5300 calBP to 5000 calBP. Posthoe diameter variation then seems to decrease from 5000 to 4800 calBP and which point the degree of variation seems to level off and remains the same through to the end of the period. Posthole diameter SD had an R value of 0.257 compared to floor space and 0.265 when compared to pithouse length.
Posthole volume variation follows a similar trend as posthole diameter, increasing in variation from 5300 to 5000 cal BP, and then shows a slow decrease until the end of the Middle Jomon. CORRELATION
These correlation matrices display all three posthole measurement correlations at once. In the left figure pithouse floor space measurements are compared to posthole SD values. In the right figure pithouse length measurements are compared to posthole SD values.
Pit, Wall, and Mirror types did not posess inner main posts, and therefore changes in posthole dimensions of those types were not able to be examined. I was able to compare posthole measures to Main and MainWall types and discovered a few clear differences between them.
Mean posthole depth was fairly stable for the Main type pithouses, seeing a possible small increase in depth towards the middle of the period, reverting back to slightly shallower depths towards the end of the Middle Jomon period. MainWall types showed a noticible decrease in posthole depth, dropping from an avereage of apx 70cm down to a bit over 60cm at apx 5000 calBP. This was followed by a gradual increase to apx 65cm deep.
The variation of posthole depths showed similat trends between the Main and MainWall types. From the start of the Middle Jomon, to apx 5000 calBP posthole depth variation within pithouses decreased slightly, which was followed by an increase in posthole depth variation.
Posthole diameter trends were also difference between the Main and MainWall types. Main posthole types saw a slight increase in posthole diameter from 5400 to 5000 calBP, and then decreased towards the end of the period. MainWall pithouse types saw a more significant increase in posthole diameters from apx 5400 calBP to 4900 calBP, and then decreased slightly towards the end of the period.
Posthole diameter variation was fairly stable for the Main types, but showed some clear changes in MainWall types. MainWall type posthole diameter variation increased from apt 5300 calBP to 5000 calBP. The degree of variation decreased from 5000 to 4800 calBP, and then show another slight increase heading towards the end of the period.
Posthole volume trends between Main and MainWall types were overall similar, but did have some differences as well. Both Main and MainWall types showed an increase towards the middle of the period and a decrease in volume after that peak. The timing of the peaks and the degree of change was slightly different however. Main types peaked at apx 5000 calBP while MainWall types peaked at around 4900 calBP. Main types also showed a more significant decrease than MainWall types did.
Posthole volume variation trended in a similar pattern as mean posthole volume for Main types. While posthole volume showed an increase from the start of the period until apx 4900 calBP and decreased slightly afterwards, posthole volume variation might have seen a slight increase in variation at the begining of the period, but afterwards remained fairly static throughout the rest of the period.
These correlation matrices display all three posthole measurement correlations at once. The figures on top compare Main and MainWall pithouse floor space measurements to posthole SD values. The figures below compare Main and MainWall pithouse length measurements to posthole SD values
These correlation matrices display all three posthole measurement correlations at once. The figures on top compare Main and MainWall pithouse floor space measurements to posthole SD values. The figures below compare Main and MainWall pithouse length measurements to posthole SD values
I have reexamined the skeletal data I gathered for my MA thesis and have applied the combination of aoristic analysis and MCMC to the data. I have included two measures, the 15p5 (left) and the 5p0 (right) ratios. The 15p5 compares subadults from 5 to 19 years old to the total population of 5 years old. The 5p0 ratio compares juveniles 0 to 5 years old to the total population. Although I included nationwide data in my MA thesis, I have focused only on the Kanto and Chubu regions in this analysis.
The plots below include several different methods of filtering data included in the analysis. 15p5 measures are on the left and 5p0 measures are on the right. The upper left corner plots removed data outside of the Middle Jomon period prior to creating the plot. The plot in the upper right corner left in data from before and after the Middle Jomon period, but only plotted the portion within the Middle Jomon period. The bottom left plot filtered data before creating the plot, but data from 1000 years before and after the Middle Jomon was included. The plots in the bottom right corner include all skeletal data, regardless of the time period. Plots including data from outside the time period in question helps to smooth the data and mitigate edge effects, however, pithouse analysis was unable to benefit from these effects as data for pithouses only included data from the Middle Jomon period.
